[all-commits] [llvm/llvm-project] 4f17b1: Fix for AMDGPU MUL_I24 known bits calculation

[jayfoad via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 4f17b1784e94609a2139e0520593884e23345f7e
      https://github.com/llvm/llvm-project/commit/4f17b1784e94609a2139e0520593884e23345f7e
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2019-12-16 (Mon, 16 Dec 2019)

  Changed paths:
    M llvm/include/llvm/Support/KnownBits.h
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    A llvm/test/CodeGen/AMDGPU/amdgpu-mul24-knownbits.ll

  Log Message:
  -----------
  Fix for AMDGPU MUL_I24 known bits calculation

Summary:
At present, the code calculating known bits of AMDGPU MUL_I24 confuses the concepts of "non-negative number" and "positive number".

In some situations, it results in incorrect code. I have a case where the optimizer replaces the result of calculating MUL_I24(-5, 0) with -8.

Reviewers: foad, arsenm

Reviewed By: arsenm

Subscribers: foad, arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits

Tags: #llvm

Patch by Eugene Kuznetsov.

Differential Revision: https://reviews.llvm.org/D70367